Purpose: compare subjective (Rx) and objective (ObjRx) refractions outcomes with two autorefractors models and an aberrometer in eyes implanted with a hydrophobic trifocal IOL (FineVision POD F GF, Physiol, Liége, Belgium) and a hydrophilic one (FineVision POD F, Physiol, Liége, Belgium).
Methods: prospective comparative cohort study, with 100 subjects randomly assigned to either the POD F group (n=50) or the POD F-GF group (n=50). Postoperative eye examinations at 1month visit included 7 result sets, one for each assessment method: Rx, AR (automated refraction measured with the autorefractor KR8800), WF-P (Zernike-coefficients-based objective refraction, photopic pupil size), WF-M (Zernike-coefficients-based objective refraction, mesopic pupil size), WF-4 (Zernike-coefficients-based objective refraction, 4 mm pupil), OPD-C (automated refraction measured with the aberrometer OPD in the central pupil/photopic conditions), and OPD-M (automated refraction measured with the aberrometer OPD under mesopic conditions).Results: Mean differences between ObjRx and Rx reached statistical significance for sphere and spherical equivalent (M) only with OPD-C in the POD F-GF group. All ObjRx methods showed significant differences with Rx in the POD F group, with some values differing by more than 0.50 D (-0.58 D in M for the WF-P). Bland Altman plots showed better agreement for astigmatic components, and for sphere and spherical equivalent in both IOL groups measured with AR and OPD-M.Conclusions: None of the objective methods of refraction evaluated in this study were as reliable as the subjective refraction, irrespective of the lens material, but POD F-GF ObjRx seem to differ less with Rx than POD F ObjRx values.